Sabayon Linux unbootable; infinitely repeating errors, desktop never loads, need suggestions to fix it

I’ve run into a bit of an issue with one of my Sabayon laptops. It’s a Lenovo T440, if that matters. I don’t remember the exact system specs, it’s from about 2015 or 2016, I think.

The machine has been running flawlessly for quite some time now, but hasn’t been rebooted in a few weeks. It has been updated regularly. Mostly it gets used by my wife for email and light web browsing, stuff like that.

Today, I discovered that the entire menu system had lost everything. I don’t know when that happened, we usually run it with six virtual desktops with all the software we normally use, loaded into the various desktops and we switch as needed. But, today, I decided a reboot was in order, I went looking for the shutdown menu, and it wasn’t there. at all. NOTHING was there. I had to switch to a console and shut it down manually from the command line.

Anyway, when I rebooted it, it starts up to Grub bootscreen normally, but then once it starts booting, it reaches a certain point, switches to the desktop loading screen… and stops. It doesn’t load the Sabayon logo, it doesn’t show a pointer, it just shows three white rectangles in a row on the screen, and just sits there. It never gets past that point, no matter how long we wait.

Switching to a console, I see an unending flow of repeating errors scrolling by almost too fast to read. The same errors, over and over and over, and it just never loads the desktop or gets past those errors. I can’t even boot it into single user mode, even trying that, it stops at these errors and does not boot past them.

I have rebooted it several times and tried to capture the errors. Below is the text of everything I could capture:

[ 0.456322] ACPI Error: Needed type [Reference], found [Integer] 00000000a8e2096e (20190816/exresop-66)

[ 0.456389] ACPI Error: AE_AML_OPERAND_TYPE, While resolving operants for [opcodeName unavailable] (20190816/dswexec-431)

[ 0.456456] ACPI Error: Aborting method _PR.CPU0._PDC due to previous error (AE_AML_OPERAND_TYPE) (20190816/psparse-529)

/lib/systemd/systemd-udevd: error while loading shared libraries: libsystemd-shared-247.so: cannot open shared object file: no such file or directory

/init: line 96: [: -gt: unary operator expected

/lib/dracut-lib.sh: line 546: [: -ge: unary operator expected

/init: line 157: [: -lt: unary operator expected

The first lines, about ACPI, show up on the screen briefly just before the endless scrolling of errors starts running, but they are not repeated. the lines about init, systemd-udevd, and dracut-lib.sh repeat infinitely. At this point I don’t know what if anything the repeating errors have to do with why it won’t load the desktop or even load a single user environment.

I don’t know what caused this and I’m not sure how best to go about fixing it. Any suggestings or recommendations are most welcome to recover the system, preferably without losing its current configuration.

Forgot to mention, it’s the KDE version of Sabayon, and I have the latest KDE Sabayon 19.03 installation DVD burned and ready to go. I’ve tested it and it boots on the T440 just fine. Very slow over the USB-connected DVD drive (the T440 does not have a built in optical drive), but it does boot and run.

I wish there was a way to do an in-place “upgrade/re-install” using the DVD, but if there is, I don’t know how to do it. I would really like to hear any and all suggestions/recommendations on how to recover this machine back to usability again.

Sorry to disappoint, I have no fix for this, my ASUS laptop started to display those ACPI errors lately, but it still boots and starts the system (also KDE). It’s UEFI. My MBR PC does not show this. Seems to happen since 2018 for many. Look here: https://forums.lenovo.com/t5/Other-Linux-Discussions/Thinkpad-T440-T450-T460-Bios-ACPI-Errors/m-p/4239132?page=1#4275991.

Okay, that covers the ACPI issues. Which, I suspect, were going on long before the system became unbootable.

Beyond that… what about making the system bootable again? Any suggestions for restoring the system to a useable state?

It looks like it is a systemd, a config, maybe a driver problem.
If you think it’s the boot, I’d use the latest Sabayon ISO, chroot, then apply the kernel-switcher tool that would rebuild the boot for you. But this is kind of long shot, I can’t tell why is this happening, so it’s just a way to know it’s something related to the kernel. I’d check the disk first too. If you use kernel-switcher, it’s possible to switch to the same kernel in order to fix a boot problem.

Thank you for that advice, that’s my next step, I think. I suspect it’s something with the initramfs, because it never gets to the point where the screen font changes and it starts listing the various steps in the load/boot process. It just dumps to a never-ending scroll of two or three failures to load and never gets past that point.

I have the latest ISO, it is booting now. I chrooted into it earlier today, and tried to run equo update, which ran, downloaded and installed three X-related files, and… that’s it.

This particular system has only ever had the original kernel installed on it, I never bothered to switch to a newer kernel, so this kernel switching thing might just be the ticket to a fresh boot. We’ll see, I’m going to try it.

I will update when I have a result.

Well, I don’t know exactly what it was, and I’m not sure the logs will show me what it was… but it’s fixed.

I could not get kernel-switcher to run in chroot, I got a long list of errors and it just would not run.

However, I was able to run equo and re-install the kernel and bzimage/initramfs… that seems to have done the trick, it would appear. The system is booting again as if nothing happened, the menus are back, and it seems to be running properly again.

Thanks for the assist, I really appreciate it. I will keep this trick up my sleeve for future use.

Glad you’ve solved it! :+1: